Claude Official Skill: canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other s...
What it is
Claude Official Skill: canvas-design is a skill for Claude Code that enables the creation of visual art in .png and .pdf formats. It uses design philosophy and programmatic rendering to generate posters, artwork, charts, and other visual outputs directly from Claude Code.
This skill targets developers and designers using Claude Code who want to generate visual assets programmatically without leaving their development environment.
How it saves time or tokens
Instead of switching to design tools like Figma or Photoshop for simple visual outputs, the canvas-design skill generates artwork directly in Claude Code. The skill includes design guidelines, color theory, and layout principles so the agent produces visually coherent results without extensive prompt engineering. Token estimate is approximately 500 tokens.
How to use
- Install the skill:
claude skill install anthropics/skills/canvas-design
- Or manually copy the SKILL.md content to
.claude/skills/canvas-design/SKILL.md.
- Ask Claude Code to create visual designs, and the skill activates automatically.
Example
# Example prompts that activate the canvas-design skill:
# 'Create a poster for our product launch'
# 'Design a social media card for this blog post'
# 'Generate an infographic showing our architecture'
# 'Make a cover image for this article in 1200x630'
Related on TokRepo
- AI Tools for Design -- Design and visual creation tools
- Prompt Library -- Curated prompts and skill rules
Common pitfalls
- The canvas-design skill generates static images. It does not create interactive or animated designs.
- Output quality depends on the specificity of your prompt. Include dimensions, color preferences, and content requirements for best results.
- Complex layouts with many elements may require multiple iterations. Start with simple designs and add complexity incrementally.
Frequently Asked Questions
The canvas-design skill produces .png and .pdf files. PNG is used for raster images like posters and social media cards. PDF is used for print-ready designs and documents.
Yes. canvas-design is part of the official Anthropic skills repository maintained by the Claude Code team. It is installed from the anthropics/skills GitHub repository.
Yes. The skill can generate charts, diagrams, infographics, and data visualizations. Provide the data and desired chart type in your prompt.
No. The skill embeds design principles including color theory, typography guidelines, and layout rules. You describe what you want in natural language and the skill applies design best practices.
Yes. Specify style preferences in your prompt such as color palette, typography style, layout orientation, and visual density. The skill adapts its output based on your instructions.
Citations (3)
- Anthropic Skills GitHub— Official Anthropic skills repository for Claude Code
- Anthropic Documentation— Claude Code documentation and skill installation
- Anthropic Design Guidelines— Design principles for visual communication
Related on TokRepo
Source & Thanks
Created by Anthropic. Licensed under MIT. anthropics/skills
Discussion
Related Assets
Claude-Flow — Multi-Agent Orchestration for Claude Code
Layers swarm and hive-mind multi-agent orchestration on top of Claude Code with 64 specialized agents, SQLite memory, and parallel execution.
ccusage — Real-Time Token Cost Tracker for Claude Code
CLI that reads ~/.claude logs and breaks down Claude Code token spend by day, session, and project — pluggable into your statusline.
SuperClaude — Workflow Framework for Claude Code
Adds 16+ slash commands, 9 cognitive personas, and a smart flag system to Claude Code in one pipx install.